Predicting the future has always been difficult in any fast-changing field and computational fluid dynamics (CFD) is no exception. But, says Milovan Peric, it seems certain it will spread to almost all areas of engineering.

Computational fluid dynamics (CFD) is undergoing rapid development and adaptation, due to both advances in hardware technology and ever increasing demands by users. There are several developing trends that can be observed from both academic and industrial points of view.
It seems indisputable that the use of CFD will spread in the future to almost all areas of engineering, biomedical, and environmental research and development. One reason for this is the increasing awareness of the benefits and potential that CFD offers. This has been due to better education about CFD at both academic and industrial levels. There is also a rising trend for use by non-experts, which CFD software authors will have to properly address by providing simple-to-learn and simple-to-use, application-tailored versions. The EZ Tools of the CD adapco Group are already pointing in that direction.

Quality of simulation

Grid-generation issues have always plagued CFD practitioners and one of the major efforts in the future will be further automation in this area. This development will be driven not only by the desire to reduce turn-around time, but also to increase the quality of simulation by a more faithful representation of the geometric details and higher grid quality.
As the optimisation margins become tighter, the issue of accuracy will play an even greater role. CFD tools will have to provide error estimates and error-guided automatic adaptation of both the grid and the solution method.
Furthermore, CFD tools will become part of a global CAE environment, interacting with other design tools to make optimisation faster, easier and more reliable.
Demand will doubtless grow for a more detailed consideration of complicated physics boundary conditions and multi-physics phenomena. These might include coupled simulation of fluid-structure interaction, phase change, electromagnetic effects or multiple phases.
Many of the processes will have to be analysed in an unsteady fashion to be more faithful to nature, which will make the use of Large-Eddy and Direct Numerical Simulations unavoidable.
The immense memory and computing time issues that hamper today's users will be improved by advancements in computer technology.

Friendly interface

Another challenge facing CFD software authors comes from the growing need for a more user-friendly graphical interface. In particular, this extends to the need for the CFD results to be presented in a visually attractive and self-explanatory way, to meet the needs of less technical users. For example, the computer animation of unsteady flow fields is commonplace today, but could be replaced by virtual reality presentations in the future.
Of course, the modelling, discretisation, and solution methods used in CFD solvers are also undergoing changes, gaining in efficiency, accuracy, and robustness. This trend is also continuing, with sophisticated coupled solvers slowly replacing their less robust sequential counterparts.

Furthermore, the need for collaboration on a broader scale between software providers like the CD adapco Group and research institutions, academia, and industry will have to be intensified.
CFD has proved itself an invaluable tool for a growing number of industry sectors, and there is every reason to suggest that this trend will continue.
Both software developers and users will be able to benefit, as companies such as The CD adapco Group strive to meet the challenges of this maturing marketplace.

Milovan Peric is with the CD adapco Group is based in Hammersmith, London. www.cd.co.uk